详细信息
混沌耗散离散粒子群算法及其在故障诊断中的应用 ( EI收录)
Chaotic dissipative particle swarm optimization and its application to fault diagnosis
文献类型:期刊文献
中文题名:混沌耗散离散粒子群算法及其在故障诊断中的应用
英文题名:Chaotic dissipative particle swarm optimization and its application to fault diagnosis
作者:王灵[1];俞金寿[1]
机构:[1]华东理工大学自动化研究所,上海200237
年份:2007
卷号:22
期号:10
起止页码:1197
中文期刊名:控制与决策
外文期刊名:Control and Decision
收录:CSTPCD;;EI(收录号:20074710936684);Scopus;北大核心:【北大核心2004】;CSCD:【CSCD2011_2012】;
基金:教育部博士点专项基金项目(20030251003)
语种:中文
中文关键词:故障诊断;粒子群优化算法;特征选择;支持向量机
外文关键词:Fault diagnosis;Particle swarm optimization;Feature selection;SVM
摘要:针对大型化工过程生产系统的高维度数据及其噪声严重影响故障诊断的性能,采用基于故障特征选择和支持向量机(SVM)的故障诊断方法.为了确保在线故障诊断的实时性和准确性,提出一种新型的混沌耗散离散粒子群(CDDPSO)算法,用于故障诊断中特征变量的搜索.仿真结果表明,CDDPSO算法能有效地搜索到全局最优解,而基于故障特征选择的故障诊断方法具有良好的故障诊断性能.
Considering the high dimensionality of data and the noises in large-scaled chemical process industry system seriously spoil the fault diagnosing performances, a fault diagnosis method based on fault feature selection and support vector machines is proposed. To ensure the real-time capability and correct rate of diagnosing online, a novel chaotic dissipative discrete particle swarm optimization (CDDPSO) is developed to search fault feature variables for diagnosis. Simulation results show that CDDPSO finds the global optima more effectively and the proposed fault diagnosis method possesses better diagnosing performance.
参考文献:
正在载入数据...
